[{"content":" The start of the RAD lab # When I joined Dr. Ambrose at Texas A\u0026amp;M as his first PhD student in 2021, the RAD Lab was little more than an empty space, just a few leftover stools and a pile of Amazon boxes.\nMy first day of grad school involved tackling this pile of boxes While we had a space at the Research Integration Center it was mainly empty highbays with few tools for robot development. As a result, RoboBall was largely designed and constructed on Texas A\u0026amp;M’s main campus.\nPhoto taken by me from the catwalk of my labmates building the first few workbenches Early testing involved loading the robot into a vehicle, driving out to the RIC, and running experiments until the batteries were depleted before returning to campus.\nAn early prototype of RoboBall in an empty highbay RAD-ical growth # Over the last four years, I’ve helped to grow the lab into a thriving research group. This experience gave me a unique perspective on how to build both technical infrastructure and collaborative culture from the ground up for developing state of the art robotic systems.\nOnce empty highbays are now a fully functional robotics workspace Leadership as a Mentor # Each summer, the RAD Lab hosts over 30 undergraduate researchers who contribute to nearly every aspect of our robotics work. I believe mentorship is a critical part of engineering and research, so I’ve volunteered to lead a student group every summer during my time in the lab.\nMy goal has always been to give students meaningful ownership over real technical problems, balancing structured guidance with the freedom to explore, make mistakes, and grow into confident contributors. This involved preparing scoped technical tasks, providing hands-on support during development and testing, and fostering an environment where questions and collaboration were encouraged.\nSummer 2024 team at the end of a feild test ","date":"25 January 2026","externalUrl":null,"permalink":"/projects/rad_lab_growth/","section":"Projects","summary":"","title":"RAD Lab Growth and Mentorship","type":"projects"},{"content":" Motivation # NASA’s Artemis rocket launched in November 2022, marking a renewed effort to return humans to the Moon. One of the key objectives of these missions is to establish a permanent base near the lunar south pole. These bases would be located close to permanently shadowed craters that may harbor frozen water, protected from direct sunlight.\nSpherical rovers could offer a unique way to explore these dark craters. Instead of relying on a slow and methodical descent of a wheeled rover, a ball-shaped rover could roll down the crater slope with relative ease (2), collect a sample (3), and then send a sample retrieval rocket back up to the crater rim, where it could be recovered by an astronaut or another rover (4).\nMechanical Design of RoboBall # RoboBall consists of two main mechanical components: an inner pendulum and an outer shell.\nRoboBall Component Part 1: Inner Mechanical Pendulum # The pendulum was designed as two mirrored halves connected by a central basket. The structural components are mirror copies of each other and include numerous mounting holes for later avionics and subsystem integration.\nElectrical Systems # Most avionics were sourced from a standard FIRST Robotics kit, with the exception of a VN-100 IMU. Later versions of the ball were retrofitted to use ros2 compatable components.\nRoboBall with FIRST sources eletrical components Testing Stands # We finished RoboBall\u0026rsquo;s pendulum within 9 months from its first design review. However we would not finalize a design for the soft outer shell until ayear afterwards. To make headway on controls and software components before the soft shell was completed, a series of testbeds were built to mimic RoboBalls rolling dynamics.\n1st Gen Testing Stand: The Hanging Stand # This stand hung the pendulum between two static beams, shown in the video. The driveshaft hex was rigidly clamped to disable relative motion between the pendulum and the ground. This stand was perfect for the initial operations tests and assembly, but its did not accurately model the reactive rolling dynamics. Your browser does not support the video tag. 2nd Gen Testing Stand: Gyro Stand # The next generation test stand used a set of flywheels to compensate for the lack of inertia of the hanging stand. These flywheels would mimic the shell’s reflected inertia to verify that the pendulum pitches upward with forward velocity. However, the belts would slip after achieving final velocity and lose active flywheel braking. This issue, in tandem with the safety risks of unsheathed flywheels, led us to scrap the stand after its functional test. Your browser does not support the video tag. 3rd Gen Testing Stand: Steering Stand # The initial hanging concepts of testing stands did not capture the rolling coupling between the shell and ground. This steering stand isolated the pendulum’s continuous drive axis from the limited steering axis. Initial balancing tests with an LQR controller were promising. Still, the difference in inertias and contact dynamics between this stand and the shell prototypes made transitioning controllers from the testbed to the robot unrealistic. Your browser does not support the video tag. 4th Gen Testing Stand: Drive Stand # The final stand was simply a set of wheels with hex adapters mounted to the outer ends of the driveshaft. This stand proved the most useful, as it served the purpose of the hanging stand to check alignments and calibrations. Simple rolling tests with the pendulum could be conducted without assembling the shell. The stand even appears in early field tests, and is the only one still used by the RoboBall team today! Your browser does not support the video tag. RoboBall Component Part 2: Soft Airtight Outer Shell # The outer shell is comprised of aluminum mounting hardware that clamps various types of soft shells. Large diameter inner and outer rings screw into each other and an outer hubcap that interfaces with the pendulum.The hubcap has an o-ring to prevent links We used this primary design to test out different iterations of soft airtight shells.\nInitial outer shell prototypes used an inner yoga ball bladder constrained by an outer nylon jacket.\nInitial Shell Iterations However, the nylon jacket was prone to tearing. To address this, I developed a molding technique using an aerosolized bedliner polymer. This paper describes the method and its advantages in tuning the outer shape of the shell.\nYour browser does not support the video tag. Assembly Timelapse # The pendulum and shell come together to complete the robot.\nYour browser does not support the video tag. ","date":"4 December 2023","externalUrl":null,"permalink":"/projects/roboball_construction/","section":"Projects","summary":"The motivation and development of RoboBall II\u0026rsquo;s Design","title":"Roboball II Pendulum and Shell Design","type":"projects"},{"content":"Since RoboBall is a multi-student project I wanted to find more approachable ways to derive a dynamic model that could be passed from student to student. Using dynamic modeling software, such as Drake, a simple urdf can yeild a numerical full dynamics model, trading lagrangian derivation for a programing exercise.\nI chose drake as a due to its open source nature, support for python, recognitioin of floating bodies, and implementation of a soft contact model. Modeling RoboBall in drake requires configuring a urdf and tuning additional dynamics not handled by a rigid body dynamics program.\nStep 1: Generate a urdf # A proper urdf can be obtained from a solidworks assembly after specifying the underlying structure of the links. For RoboBall, that looks like a floating joint to the pitch center then two branches: one to the pendulum under gravity and another to the outer soft shell.\nA diagram showing the urdf topology Step 2: Account for Friction in the Gears and Outer Shell # Drake\u0026rsquo;s urdf parser only handles the rigid body components of the system. Any losses in the system must be measured and added to the model. The following picture shows a diagram of the frictional effects an outer shell models and where they feed into the drake model.\nModels were tuned experimentally with the methods and results published in this RA-L paper.\nScreenrecord of URDF in Drake (with the soft-body model) # The completed model can be viewed in the Meshcat window.\nYour browser does not support the video tag. Step 3: Set up the ros2 environment # The simulation environment came after RoboBall had already been converted to ros2. So the same control logic on the ball can be fed against simulated data instead of sensors on the robot. An accurate digital twin would aid in tuning gains or preparing for dangerous testing environments without risking the hardware.\nOn the robot the control logic runs on an onboard jetson nano in a docker container. For the simulation, the same code is configured to run locally on the laptop.\nThe simulation runs in a standalone ros node. The compute heavy simulation setup is run on the node instantiation. So only the integrator is stepped at every callback. The predicted IMU and encoder data is sent to the control code and recieves the desired command drive and steer action.\nResults # As a result we could drive the virtual robot using the existing control logic from the hardware. The robot is not as smooth in the video because the gains are the same ones used on the robot.\nUnfotunately the drawback of working in simulation is that I must model everything, even things we do not fully understand. Whereas in a physical system mother nature will do that for you.\nYour browser does not support the video tag. ","date":"4 June 2025","externalUrl":null,"permalink":"/projects/roboball_modeling/","section":"Projects","summary":"A description on how to model RoboBall in pyDrake","title":"Roboball: Modeling with Drake and ros2","type":"projects"},{"content":" The Problem # Tests with RoboBall often ended with the system wobbling until it eventually started flipping end over end. Tests with more robust control systems only pushed this mode of instability to occur at higher speeds but would not remove it. This paper shows that the answer lies in a classic phenomenon from spacecraft dynamics that has gone largely unnoticed in ground robotics.\nThe Core Idea # High-speed spherical robots often develop a growing wobble that can escalate into end-over-end flipping. The paper shows that this behavior is not a control bug or modeling artifact, but a fundamental dynamic effect caused by the robot’s inertia interacting with rolling constraints. Specifically, rolling spherical robots with oblate inertial profiles experience a relaxation effect closely related to the Intermediate Axis Theorem (also known as the tennis-racket or Dzhanibekov effect). Under dissipation, rotating bodies naturally reorient their angular momentum toward their major inertial axis. For rolling robots, this manifests as a gradual shift from stable forward rolling into unstable “hubcap-to-hubcap” motion.\nYour browser does not support the video tag. Relaxation dynamics are well understood in tumbling asteroids but have never been systematically applied to rolling systems with contact constraints.\nThis paper: # Extends classical relaxation theory to spherical robots rolling on the ground Shows that rolling constraints couple rotational and translational energy, creating an effective dissipation mechanism Backs up the claim by conducting experiments on solid ground and water to chance the contact dynamics. Key findings from the study: # Free oblate bodies are stable, but once rolling constraints are introduced, stability changes fundamentally. Translational rolling acts like structural dissipation in asteroids, driving momentum realignment. Experiments with an empty rolling shell, a constrained pendulum, and tests on land versus water confirm the theoretical predictions. Looser constraints (e.g., rolling in water with slip) dramatically reduce or eliminate the instability. Impact on Future Spherical Robot Design # Many spherical robot models assume uniform inertia, use decoupled planar dynamics, or ignore 3D rotational effects for simplicity. My results show that inertia shape and rolling constraints fundamentally limit high-speed stability, regardless of control quality.\nRobot designers have two practical paths forward:\nInertial design Configure the shell so rolling occurs about the major inertia axis, which naturally stabilizes motion.\nControl Design Incorperating the relaxation angle with an appropriate internal actuation method could make the system aware of the effect and actively cancel it. Note that the control effort is proportional to speed so flywheels may be a better option than RoboBall\u0026rsquo;s torque constrained pendulum.\nTL;DR # This work bridges satellite attitude dynamics and ground robotics, offering a compact framework to understand inertial effects in spherical robot performance and guide both mechanical design and control strategies for future robots.\nThis paper was recently accepted for presentation at ICRA 2026. This link is for a preprint, the final version will be available shortly.\nDownload the PDF\n","date":"4 November 2025","externalUrl":null,"permalink":"/projects/roboball_and_asteroids/","section":"Projects","summary":"","title":"Asteroid Dynamics and Rolling Robots","type":"projects"},{"content":" Mobility Studies # While RoboBall was originally designed to head into craters, its unique form factor lends it to mobility in other situations.\nRoboBall on Slopes # RoboBall has a limited ability to climb slopes. By positioning the system’s center of mass as close to the outer edge of the sphere as possible, the maximum slope-climbing angle can be increased. To achieve this, we replaced the bottom ballast plate with a heavier version made of copper, improving the robot’s climbing performance.\nYour browser does not support the video tag. My labmate Rishi took these test results to showcase the improved slope climbing ability of the 6ft diameter RoboBall: paper link\nRoboBall in Water # Because the robot is inflatable, it can seamlessly transition from land to water without additional modifications.\nYour browser does not support the video tag. At higher speeds, however, RoboBall produces a “rooster tail” effect that reduces forward thrust in water. We tested this by comparing drive motor encoder data with forward speeds measured from video analysis. The figure below shows snapshots from tests conducted at increasing speeds and rooster tail magnitudes.\nNotice that at the highest speeds—where the rooster tail is largest—the forward velocity actually decreases. This effect is primarily due to a rocket equation phenomenon, where mass (water) is being ejected in the wrong direction.\n","date":"4 May 2024","externalUrl":null,"permalink":"/projects/roboball_mobility/","section":"Projects","summary":"Case Studies of RoboBall on Water and Slopes","title":"Roboball Mobility Studies","type":"projects"},{"content":" Motivation # In my last semester of undergrad, I enrolled in MEEN 408/612: Robotic Manipulators with Dr. Dharba. It was a challenging but fascinating course, and a big reason why I decided to pursue robotics as a career.\nPart of the class focused on different control strategies for manipulator arms. Since there were no lab sessions, I wanted to take what I learned and apply it to real hardware.\nOpen Source Robot Arm # The most cost-effective way to get an arm was to build one myself. I found a small group called Annin Robotics that sold pre-designed kits. These kits included machined parts and a bill of materials for sourcing the rest from Amazon or McMaster-Carr.\nDesign Lessons # The parts arrived, and I started assembly around the same time we were designing the RoboBall pendulum. As you can imagine, working with Dr. Ambrose on a custom robot is wildly different from assembling a cheap kit. Here are a few lessons I learned while building this arm alongside RoboBall:\nStandardize your fasteners – This kit used multiple bolt heads, set screws, and thread types. Keeping track of them was a pain compared to RoboBall, which only used 4-40s and 10-32s. Use standard components wherever possible – For some reason, all the screws were metric, which made it harder to find spares locally. Working in an apartment is very limiting – Building an arm requires a surprising amount of space and tools. My bedroom couldn’t function as both a workshop and a place to sleep. Design with maintenance in mind – The kit called for many components to be soldered directly together. Adding connectors might have increased bulk but would’ve saved a lot of headaches during repairs. Keep software modular and familiar – The included control software used a custom serial protocol, with all the low-level logic lumped into one massive for loop. It was nearly impossible to read or debug. Cramped workspace in my apartment Calibration Video # I eventually finished assembling the arm according to the kit instructions, but didn’t have time to go further as the RoboBall project was ramping up.\nYour browser does not support the video tag. Update — October 2025 # I’ve since finished grad school and learned a lot more about robot programming. Once I’m settled into a new job, I plan to rewrite the control software in ROS 2 and possibly implement some controllers using the pydrake library.\n","date":"2 January 2022","externalUrl":null,"permalink":"/projects/reddit-arm/","section":"Projects","summary":"Built a 6 degree-of-freedom stepper driven arm, honestly not a great design","title":"Personal Project: Six DOF Stepper Arm","type":"projects"},{"content":"RoboBall was designed to house an internal pressure control system using the FRC pneumatics kit. We were able to fit three standard air tanks, a solenoid, and a compressor within the system’s pendulum assembly.\nThe pnuematic system on it own and packaged with the rest of the avionics By venting pressure from the tanks, we could increase the ball’s internal pressure. Conversely, the compressor could pull air from the ball into the tanks, lowering the internal pressure.\nWe characterized leak and flow rates for the critical components and used this data to construct a model of the system. A detailed description is published in this paper, but the system behavior can be summarized by the following diagram.\n","date":"4 November 2025","externalUrl":null,"permalink":"/projects/roboball_pressure_control_sys/","section":"Projects","summary":"The motivation and development of RoboBall II\u0026rsquo;s Design","title":"RoboBall's Pressure Control System","type":"projects"},{"content":" A roboticist with 4 years of experience across every layer of a modern robotic system. # From autonomous semi-trucks and robot arms to unique mechatronic systems, I bring a strong blend of mechanical design, software intuition, and rigorous field testing experience to any team.\nRoboBall (bottom), an inflatable spherical robot that I developed to support my doctoral research Throughout my time with the RAD Lab, I’ve helped build an environment where ideas move quickly from whiteboard discussions to tested prototypes. I’ve learned that the best engineering outcomes come from listening carefully, communicating clearly, and giving teammates the autonomy and support they need to do their best work.\nI am no stranger to disruptive research, whether that means challenging assumptions in existing system designs, building new modeling and simulation tools from the ground up, or validating new systems through robust experimentation.\nI’m motivated by challenges that don’t have clean solutions yet, and I’m excited to bring the rigor of my academic and professional training to new problems outside of the RAD Lab. I\u0026rsquo;m looking for teams willing to do the hard, collaborative work required to build robots that truly push the state of the art.\nIf you or someone you know is looking for a roboticist like me, let\u0026rsquo;s get in touch!\nEmail Me! Browse a few highlights below, or take a deeper dive into what I have built over the years. # More Projects! Resume (PDF) My GitHub Google Scholar Work in progress # Here’s a brief write-up of what’s currently cooking in my garage:\nCurrent Project: Retrofit Stepper Arm with CAN bus \u0026middot; Work in Progress Stepper Arm ","date":"11 February 2026","externalUrl":null,"permalink":"/","section":"","summary":"","title":"","type":"page"},{"content":" Work in Progress This is an active project. I\u0026rsquo;ll keep this page updated with current status and future plans as often as I can. or check out the github repo Context # I built a stepper driven robot arm at the end of my undergraduate program. It had a number of problems and has been collecting dust for the past few years.\nNow that I’m job searching, I figured this would be a good opportunity to revisit the design and retrofit it with a more robust electronic interface.\nPersonal Project: Six DOF Stepper Arm \u0026middot;\u0026middot; Hardware Design Robot Arms Stepper Arm The Problem # The original arm was designed to be as inexpensive as possible. To minimize hardware cost, all control logic was centralized on a single microcontroller. Command parsing, sensor sampling, and stepper control logic were handled in a single large Arduino sketch. While this worked, this strategy created large wire bundle that ran from each joint down to the robot\u0026rsquo;s separate control box.\nThis bulky harness complicated assembly and reduced modularity; both in software and future hardware expansions.\nPlanned Solution # I\u0026rsquo;d like to simplify this implementation by distributing control to each joint. By equiping each joint with a local controller I believe I can reduce the amount of cables run along the arms length to two CAN wires and power lines.\nTestbed approach # I will take a testbed approach, where I\u0026rsquo;ll develop the needed firmware on an isolated testbed. Then port everything onto the robot once I\u0026rsquo;m satisfied with it.\nYour browser does not support the video tag. Design notebook # I prefer sketching ideas in powerpoint (a.k.a. ppt CAD), here is peek into my notes, they will develop as the projects matures ","date":"11 February 2026","externalUrl":null,"permalink":"/projects/can_stepper/","section":"Projects","summary":"","title":"Current Project: Retrofit Stepper Arm with CAN bus","type":"projects"},{"content":" A few highlights of my past work ","date":"11 February 2026","externalUrl":null,"permalink":"/projects/","section":"Projects","summary":"","title":"Projects","type":"projects"},{"content":"","date":"11 February 2026","externalUrl":null,"permalink":"/tags/stepper-arm/","section":"Tags","summary":"","title":"Stepper Arm","type":"tags"},{"content":"","date":"11 February 2026","externalUrl":null,"permalink":"/tags/","section":"Tags","summary":"","title":"Tags","type":"tags"},{"content":"","date":"11 February 2026","externalUrl":null,"permalink":"/tags/work-in-progress/","section":"Tags","summary":"","title":"Work in Progress","type":"tags"},{"content":"","date":"25 January 2026","externalUrl":null,"permalink":"/tags/leadership/","section":"Tags","summary":"","title":"Leadership","type":"tags"},{"content":"","date":"25 January 2026","externalUrl":null,"permalink":"/tags/mentorship/","section":"Tags","summary":"","title":"Mentorship","type":"tags"},{"content":"","date":"25 January 2026","externalUrl":null,"permalink":"/tags/team-building/","section":"Tags","summary":"","title":"Team Building","type":"tags"},{"content":"","date":"4 November 2025","externalUrl":null,"permalink":"/tags/dynamics/","section":"Tags","summary":"","title":"Dynamics","type":"tags"},{"content":"","date":"4 November 2025","externalUrl":null,"permalink":"/tags/icra-2026/","section":"Tags","summary":"","title":"ICRA 2026","type":"tags"},{"content":"","date":"4 November 2025","externalUrl":null,"permalink":"/tags/roboball/","section":"Tags","summary":"","title":"RoboBall","type":"tags"},{"content":"","date":"4 November 2025","externalUrl":null,"permalink":"/tags/control-systems/","section":"Tags","summary":"","title":"Control Systems","type":"tags"},{"content":"","date":"4 November 2025","externalUrl":null,"permalink":"/tags/hardware-design/","section":"Tags","summary":"","title":"Hardware Design","type":"tags"},{"content":"","date":"4 June 2025","externalUrl":null,"permalink":"/tags/control/","section":"Tags","summary":"","title":"Control","type":"tags"},{"content":"","date":"4 June 2025","externalUrl":null,"permalink":"/tags/pydrake/","section":"Tags","summary":"","title":"PyDrake","type":"tags"},{"content":"","date":"4 June 2025","externalUrl":null,"permalink":"/tags/ros2/","section":"Tags","summary":"","title":"Ros2","type":"tags"},{"content":"","date":"4 June 2025","externalUrl":null,"permalink":"/tags/simulation/","section":"Tags","summary":"","title":"Simulation","type":"tags"},{"content":"","date":"4 June 2025","externalUrl":null,"permalink":"/tags/urdf/","section":"Tags","summary":"","title":"URDF","type":"tags"},{"content":"","date":"4 May 2024","externalUrl":null,"permalink":"/tags/field-testing/","section":"Tags","summary":"","title":"Field Testing","type":"tags"},{"content":"","date":"4 December 2023","externalUrl":null,"permalink":"/tags/electrical-design/","section":"Tags","summary":"","title":"Electrical Design","type":"tags"},{"content":"","date":"4 December 2023","externalUrl":null,"permalink":"/tags/prototyping/","section":"Tags","summary":"","title":"Prototyping","type":"tags"},{"content":"","date":"4 December 2023","externalUrl":null,"permalink":"/tags/test-beds/","section":"Tags","summary":"","title":"Test Beds","type":"tags"},{"content":"","date":"2 January 2022","externalUrl":null,"permalink":"/tags/robot-arms/","section":"Tags","summary":"","title":"Robot Arms","type":"tags"},{"content":"For my senior capstone, I was part of a team of six mechanical engineering students to improve a hard-connect towbar system for semi-truck platooning.\nMotivation # The project was part of a larger effort to explore the hardware and control aspects of semi-autonomous truck platoons. In a truck platoon, the lead truck has a driver, while following trucks rely on physical or software coupling to coordinate movement.\nOur approach used a compliant towbar between the lead and follower vehicles to physically transfer position information from the lead vehicle to the follower. The first version of this design, created by a previous senior design team, was massively overbuilt for the expected loads and only allowed compliance in compression.\nPrevious Design # Initial Prototypes # We spent a majority of the year looking into a collapsable ratcheting system. It showed promise but had serious safety concerns. With three months left we abandoned that radical idea to improve the existing design.\nFinal Design Goals # My Role at this point was coordinating my teamembers in various subteams. I worked with our main point of contact on the redesign to determine simple design goals we could hit by the end of the semester.\nWe sought to introduce bi-directional compliance, add an extension system to simplify loading, and lighten the overall weight of the towbar.\nI separated the team into two groups, one to work on the front extension system (left in the below picture) and another to repurpose the springs from the old design into a geometry that allowed the two way compliance (right). By dividing the work amoungst the team and letting them have individual ownership of the project I focused on assisting each as needed and handling part procurement and finding a space to build.\nWith this strategy we took this design from CAD-pavement in a little over a month while in the middle of senior year classes.\nYour browser does not support the video tag. Results and Recognition # Our work earned first prize in the Fall 2021 Engineering Project Showcase where we took home a cash prize.\nThe prototype proved robust enough to continue testing even after our team graduated. You can read more about the related company and its work in this article.\n","date":"25 December 2021","externalUrl":null,"permalink":"/projects/truckplatoon/","section":"Projects","summary":"Led a senior design team improving a compliant towbar for semi-truck platooning.","title":"Undergrad Capstone: Hard Truck Platooning","type":"projects"},{"content":"","date":"6 December 2020","externalUrl":null,"permalink":"/tags/dynamic-testing/","section":"Tags","summary":"","title":"Dynamic Testing","type":"tags"},{"content":" BakerRisk Overview # BakerRisk is a consulting company based in San Antonio that specializes in hazard and risk assessment, including equipment failure, toxic material spills, structural response, and explosion hazards. Clients across industries turn to BakerRisk for process safety advice and for studying how hazards affect their operations and products.\nTo support this work, BakerRisk operates two dedicated testing facilities where full-scale experiments are conducted each year. These tests generate data used both to develop safety procedures for hazardous materials and to evaluate how structures respond under extreme conditions. Testing is broadly divided into three groups: Process Safety, Structures, and Blast. Process Safety focuses on mitigation procedures, Structures investigates building response and design, and the Blast group studies explosion phenomena and is where I spent most of my time.\nDuring my internship, I split time between two test sites:\nWilfred E. Baker (WEB) site — east of San Antonio, home to the Shock Tube. Box Canyon Test Facility — west of Uvalde, home to the Deflagration Load Generator (DLG). The DLG simulates pressure waves from vapor cloud deflagrations, while the Shock Tube produces controlled air-blast loads with a high-pressure driver. Both are used to push safety-critical structures and materials to their limits.\nBox Canyon Projects # The Deflagration Load Generator (DLG) # The DLG is a combustion-driven pressure wave generator: an airtight steel chamber with one open end facing the test pad. By varying internal congestion (pipes or obstacles), flame speeds and blast loads can be controlled. Higher congestion produces faster flame fronts and stronger pressure waves.\nGas sampling ports within the DLG monitor the fuel–air mixture, while plastic sheeting seals the opening before ignition. Tests are triggered remotely, blasting whatever specimen is on the pad.\nDLG Testing Group 1 — Blast Resistant Module \u0026amp; Scaffolding # My first project was testing an anchored Blast Resistant Module (BRM). The goal was to characterize its response to explosions on a work site. Previous tests had used an unanchored BRM that slid across the pad; this time, the rectangular module was fixed to the ground with six 1-inch bolts. We recreated the same loads to compare anchored vs. unanchored responses.\nTwelve pressure gauges were mounted on each non-loaded face with two inside and five on the loaded face.\nTo maximize efficiency, the BRM tests were combined with a separate study on standard scaffolding structures commonly used as temporary shelters. Both test setups shared the same DLG platform.\nDLG Testing Group 2 — Baseline Tests # Without internal congestion, the flame cannot accelerate into a detonation, and the DLG produces only a large fireball—seen in the video below. This run is used by the test engineers to calibrate the response of the system without the congestion.\nYour browser does not support the video tag. Most of my work with the DLG involved working independently in the field with one other intern, Tyler. We communicated with the full-time engineers over radio and were reponsible to the setup and calibration of the pressure guages mounted throughout the test pad, checking dessicant levels in the fuel delivery system, and maintenance of the test abort chord (a simple remote winch that would tear the plastic to vent any contained gas).\nWEB Projects — The Shock Tube # The Shock Tube subjects samples to precisely controlled blast pressures and impulses. Unlike the DLG, it uses compressed air rather than combustion.\nIt consists of three main sections:\nDriver — pressurized air chamber. Spool — buffer section. Vent — open-ended test section. The sections are separated by thin aluminum foils that rupture at specific pressure differences (e.g., 0.018-in foil breaks at ~20 psi). By stacking foils and pre-pressurizing the spool, precise load conditions can be generated. Once the foils rupture, the load propagates into the vent and onto the specimen.\nBallistic Glass Testing # One major program focused on the response of treated and laminated glass of varying thickness under different shock loads. Glass panes were mounted in a dual nested steel frame held against the shock tube. The frame was unbolted, allowing slight movement so that load cells could record edge forces on the glass.\nMy role in this project involved wiring the window frames with load cells and slow-motion video cameras. Even after flying glass would occasionally cut the wires. Additionally, Tyler and I assissted full time staff with the repeated trials of preparing and testing the windows.\nFortress Wall Pouring \u0026amp; Testing # This program investigated reinforced wall sections designed for high-blast environments. Test panels were poured, cured, and subjected to shock loads at WEB to assess their resistance.\nMy role in this involved tying the internal rebar frame together and pouring the concrete required. I\u0026rsquo;m glad I\u0026rsquo;m not a civil engineer.\n","date":"6 December 2020","externalUrl":null,"permalink":"/projects/bakerrisk/","section":"Projects","summary":"An overview of my time as a testing intern at BakerRisk in San Antonio from Aug–Dec 2020.","title":"Intern Projects: BakerRisk","type":"projects"},{"content":"","date":"6 December 2020","externalUrl":null,"permalink":"/tags/structures/","section":"Tags","summary":"","title":"Structures","type":"tags"},{"content":" Motivation # Engineers use mechanical models of soft tissue to design prosthetics or synthetic grafts that can seamlessly mesh with the patients skin.\nHowever these mechanical models are complicated, and constructing experiments to accuratly replicate the boundary conditions the model requires special consideration. As part of a larger experimental program to characterize the bi-axial stress response of soft tissues, my work involved desiging a tensile clamp that would mitigate the stress sheilded regions of rigid clamps perpendicular to the stress direction.\nThe attached final poster presents my work on the design of a fishook pulley system to transfer tensile loads to the specimen that avoids the stress sheilding effects of rigid clamps. The poster was presented at a final public research symposium at the end of the semester.\nWhile this was my first introduction to research and I enjoyed the process, I learned that I do not have the stomach to handle organic components for a living.\n","date":"24 April 2019","externalUrl":null,"permalink":"/projects/bmel/","section":"Projects","summary":"I helped deign a suture-fishook clamping mechanism to bi-axially tensile test soft tissues.","title":"Undergrad Research: Bi-Axial Tensile Testing for Soft Tissues","type":"projects"},{"content":"","externalUrl":null,"permalink":"/authors/","section":"Authors","summary":"","title":"Authors","type":"authors"},{"content":"","externalUrl":null,"permalink":"/categories/","section":"Categories","summary":"","title":"Categories","type":"categories"},{"content":"","externalUrl":null,"permalink":"/roboball/","section":"Roboball","summary":"","title":"Roboball","type":"roboball"},{"content":"","externalUrl":null,"permalink":"/series/","section":"Series","summary":"","title":"Series","type":"series"}]